Comments (4)
The operation of two separate convolutions in the code is identical to a convolution firstly and then followed by splitting.
The only difference is that now it requires two different BN layers rather than only one.
from scnet.
Actually, this two implementations have different computational complexity. BTW, I wonder that the computational complexity and accuracy reported in the paper is based on which implementation? Thanks.
from scnet.
This is a misunderstanding. In Figure.2, the 1x1 convolutions before and after the 3x3 convolution in the bottleneck structure are omitted. Self-calibrated convolution is to replace the 3x3 convolution. Thus the implementation is identical to the paper and the computational complexities are roughly the same. The computational complexity and accuracy reported in the paper are obtained with exactly the same code we released here.
from scnet.
Okay, I see. You mean this implementation splits the first 1x1 convolution in the original bottleneck into two 1x1 convolutions. In this way, two BN+ReLU layers are required. Thanks for your reply and now it's clear to me.
Besides, if someone want to apply SCConv to other structures without pre-convolution layer, feature splitting would be a more general way to do it, right?
from scnet.
Related Issues (20)
- SCNeXt code available?
- Could you share how you visualize the feature maps in the papar HOT 1
- How to split feature map C to C/2?
- How to deal with when output channel number C‘ are not equal to input channel number C? HOT 1
- 数据集版本 HOT 3
- How to solve the problems when inplanes != planes
- 这个是只有参数吗?网络在哪里去找呀 HOT 1
- about adding SCConv into resnet20
- is there a better way to add SCCONV in resnet18 HOT 2
- the code for depth completion.
- 将SCBottleneck应用到depth completion
- SC-Conv Implementation on Keras
- 替换Resnet
- How do other networks use SCConv convolution block? HOT 3
- why
- Are you provide the full training code in the futrue?Want see the diff with other model training details HOT 1
- Could you help me? HOT 2
- Expected more than 1 value per channel when training HOT 1
- Any improvement if SCNet is applied to anchor-free architecture of object detection method? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from scnet.